The increase in computing power, spatial densities in semiconductor based devices and energy efficiency of the same allow for ever more efficient and small microelectronic sensors, processors and other machines. These have found wide use in mobile and wireless applications and other industrial, military, medical and consumer products.
Even though computing energy efficiency is improving over time, the total amount of energy used by computers of all types is on the rise. Hence, there is a need for even greater energy efficiency. Most efforts to improve the energy efficiency of microelectronic devices have been at the chip and transistor level, including with respect to transistor gate width. However, these methods are limited and other approaches are necessary to increase device density, processing power and to reduce power consumption and heat generation in the same.
One field that can benefit from the above improvements is in switched inductor power conversion devices. Power supplies include power converters that convert one form of electrical energy to another. A regulated power supply is one that controls the output voltage or current to a specific value; the controlled value is held nearly constant despite variations in either load current or the voltage supplied by the power supply's energy source.
Power converters for electronic devices can be broadly divided into AC-AC, AC-DC and DC-DC power converters. Each of these classes use similar devices, techniques and topologies as the others. Modern integrated circuits using advanced CMOS technology will run on power supplies with voltages at 1V-DC or less, while the power levels delivered to a computer are typically at 120V-AC or higher. The 120V-AC is provided by the grid, where the 120V-AC is derived using AC-AC converters from much higher voltage levels for power transmission. Once delivered to the computer, the 120V-AC power is down-converted in the computer to 1V-DC for the microprocessor through a series of power converters, AC-DC converters will generally provide a range of DC voltages such as 3.3V, 5V and 12V, and then a buck converter will take one of those power levels and down-convert to the 1V-DC required by the microprocessor.
AC-AC, AC-DC and DC-DC converters can be further divided into line-frequency (also called “conventional” or “linear”) and switching power supplies. Conventional AC-AC and AC-DC power supplies are usually a relatively simple design, but they become increasingly bulky and heavy for high-current equipment. This is due to the need for large mains-frequency transformers and heat-sinked electronic regulation circuitry. Conventional DC-DC converter, linear voltage regulators, produce regulated output voltage by means of an active voltage divider that consumes energy, thus making efficiency low.
A switched-mode power supply of the same rating as a conventional power supply maintains a smaller footprint with better efficiency but at the expense of being more complex. In an AC-AC switched-mode power supply (SMPS), the AC mains input is directly rectified and then filtered to obtain a DC voltage. The resulting DC voltage is then switched on and off at a high frequency by electronic switching circuitry, thus producing an AC current that will pass through a high-frequency transformer or inductor. In a DC-DC SMPS, a DC input voltage is switched on and off at a high frequency by electronic switching circuitry and then passed through a transformer or inductor, where the output of the transformer or inductor is connected to a decoupling capacitor. The output of the inductor or transformer is the converted DC power supply.
Switching occurs at a very high frequency (typically 10 kHz-500 MHz), thereby enabling the use of transformers and filter capacitors that are much smaller, lighter, and less expensive than those found in linear power supplies operating at mains frequency.
Switched-mode power supplies are usually regulated, and to keep the output voltage constant, the power supply employs a feedback controller that monitors current drawn by the load. The switching duty cycle increases as power output requirements increase which puts increasing demands on the constituent components, particularly the inductors. Switch-mode power supplies also use filters or additional switching stages to improve the waveform of the current taken from the input power source. This adds to the circuit complexity, with the inclusion of additional inductors and capacitors.
Additionally, the delivery of low voltage/high current power is also challenging because power loss increases with higher currents, as follows:Ploss=I2Rwhere, Ploss is the power loss over the length of wire and circuit trace, I is the current and R is the inherent resistance over the length of wire and circuit trace. As such, and to increase overall performance, there has been a recognized need in the art for large scale integration of compact and dense electrical components at the chip level, such as, for use with the fabrication of complementary metal oxide semiconductors (CMOS).
With the development of highly integrated electronic systems that consume large amounts of electricity in very small areas, the need arises for new technologies which enable improved energy efficiency and power management for future integrated systems. Integrated power conversion is a promising potential solution as power can be delivered to integrated circuits at higher voltage levels and lower current levels. That is, integrated power conversion allows for step down voltage converters to be disposed in close proximity to transistor elements.
Unfortunately, practical integrated inductors that are capable of efficiently carrying large current levels for switched-inductor power conversion are not available. Specifically, inductors that are characterized by high inductance (>1 nH), low resistance (<1 Ohm), high maximum current rating (>100 mA), and high frequency response whereby no inductance decrease for alternating current (AC) input signal greater than 1 MHz are unavailable or impractical using present technologies.
Furthermore, all of these properties must be economically achieved in a small area, typically less than 1 mm2, a form required for CMOS integration either monolithically or by 3D or 2.5D chip stacking. Thus, an inductor with the aforementioned properties is necessary in order to implement integrated power conversion with high energy efficiency and low output voltage ripple which engenders periodic noise in the output of the converter's output.
The use of high permeability, low coercivity material is typically required to achieve the desired properties on a small scale. In electromagnetism, permeability is the measure of the ability of a material to support the formation of a magnetic field within itself. In other words, it is the degree of magnetization that a material obtains in response to an applied magnetic field. A high permeability denotes a material achieving a high level of magnetization for a small applied magnetic field.
Coercivity, also called the coercive field or force, is a measure of a ferromagnetic or ferroelectric material to withstand an external magnetic or electric field. Coercivity is the measure of hysteresis observed in the relationship between applied magnetic field and magnetization. The coercivity is defined as the applied magnetic field strength necessary to reduce the magnetization to zero after the magnetization of the sample has reached saturation. Thus coercivity measures the resistance of a ferromagnetic material to becoming demagnetized. Ferromagnetic materials with high coercivity are called magnetically hard materials, and are used to make permanent magnets.
Coercivity is determined by measuring the width of the hysteresis loop observed in the relationship between applied magnetic field and magnetization. Hysteresis is the dependence of a system not only on its current environment but also on its past environment. This dependence arises because the system can be in more than one internal state. When an external magnetic field is applied to a ferromagnet such as iron, the atomic dipoles align themselves with it. Even when the field is removed, part of the alignment will be retained: the material has become magnetized. Once magnetized, the magnet will stay magnetized indefinitely. To demagnetize it requires heat or a magnetic field in the opposite direction.
High quality inductors are typically made from high permeability, low coercivity material. However, high permeability materials tend to saturate when biased by a large direct current (DC) magnetic field. Magnetic saturation can have adverse effects as it results in reduced permeability and consequently reduced inductance.
Soft ferromagnetic materials have a number of useful applications within circuits and microelectronic applications. High permeability and low coercivity are two properties that are useful for enhancing inductance. Inductance is a physical phenomenon that can be explained by the combination of Oersted's law (an electric current in a conductor creates a proportional magnetic field) and Faraday's law (a time varying magnetic flux induces an electric potential in nearby conductors). The consequence of inductance is that a change in electric current through a conductor will result in an induced electric potential (EMF) that resists the change in current. Soft magnetic materials exhibit a high permeability and consequently can be placed proximal to conductors within the path of magnetic fields that originate from these conductors, in order increase inductance values.
Typically, within the plane of the film there exists a hard axis of magnetization and an easy axis of magnetization. Along the easy axis, the material tends to exhibit greater coercivity and a highly non-linear relationship between applied magnetic field and magnetization. Along the hard axis, the material tends to exhibit lower coercivity and a relatively linear relationship between applied magnetic field and magnetization.
FIG. 1 illustrates a top view of a toroidal inductor 10 according to the prior art. The inductor 10 includes an annular magnetic core 110 and an inductor coil 120. The coil 120 wraps around the core 110 and extends in a circular direction with respect a core plane 125 that bisects the core 110. The inductor 10 generates a closed loop magnetic field 140 parallel to the circular direction of the core. As illustrated, the magnetic field 140 induces the core 110 to form a hard axis 150 and an easy axis 160 in the plane 125, with the hard axis 150 orthogonal to the easy axis 160. Thus, the magnetic field 140 passes through about half the core 110 in general alignment with the hard axis 150 and about half the core 110 in general alignment with the easy axis 160. This is undesirable because the easy axis 160 of the core has a greater coercivity (and thus non-linearity), which results in magnetic saturation as discussed above.
Accordingly, there is a need for high quality inductors to be used in large scale CMOS integration. This provides a platform for the advancement of systems comprising highly granular dynamic voltage and frequency scaling as well as augmented energy efficiency. The present disclosure contemplates the novel fabrication of efficient and compact on-chip inductors and practical methods for manufacturing operating thereof for remedying these and/or other associated problems.